---
title: "Auto Roleplay"
description: "A high-intelligence, long-context model with low input cost for repeatedly sent conversations."
canonical_url: "https://minirouter.sh/docs/routers/auto-roleplay"
markdown_url: "https://minirouter.sh/docs/routers/auto-roleplay.md"
last_updated: "2026-09-26"
---

# Auto Roleplay

A high-intelligence, long-context model with low input cost for repeatedly sent conversations.

## Call it

```sh
curl https://api.minirouter.sh/v1/chat/completions \
  -H "Authorization: Bearer $MINIROUTER_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "minirouter/auto:roleplay",
  "messages": [
    {
      "role": "user",
      "content": "Explain an API gateway in one sentence."
    }
  ],
  "max_tokens": 256
}'
```

Examples read your key from `MINIROUTER_KEY`.

## How it picks

Best first. The first model that passes answers.

1. [zai/glm-5.3-flash](https://minirouter.sh/models/zai/glm-5.3-flash) · first available
2. [openai/gpt-6-luna](https://minirouter.sh/models/openai/gpt-6-luna) · available
3. [deepseek/deepseek-v4-flash-0731](https://minirouter.sh/models/deepseek/deepseek-v4-flash-0731) · available
4. [google/gemini-3.8-flash](https://minirouter.sh/models/google/gemini-3.8-flash) · available

Live availability: [Auto models](https://minirouter.sh/models/auto).

Allowed on your key, within your cost tier, healthy, and able to take the request.

## Pricing

- **Price**: Provider cost plus 8%. Named in `x-minirouter-model`. You pay for the model that answered.
- **Bound**: At most $0.81 in and $4.05 out per million tokens. The dearest model in the list, at this router's rate.